Semi-Supervised Active Learning for Semantic Segmentation in Unknown Environments Using Informative Path Planning.
Julius RückinFederico MagistriCyrill StachnissMarija PopovicPublished in: IEEE Robotics Autom. Lett. (2024)
Keyphrases
- path planning
- unknown environments
- semantic segmentation
- semi supervised
- active learning
- weakly supervised
- mobile robot
- obstacle avoidance
- semi supervised learning
- conditional random fields
- labeled data
- superpixels
- unlabeled data
- dynamic environments
- pairwise
- supervised learning
- motion planning
- multi robot
- learning algorithm
- training examples
- multi view
- learning process
- training set
- object categories
- degrees of freedom
- indoor environments
- autonomous navigation
- machine learning
- segmentation method
- mobile robotics
- free space
- multiple robots